sxk000

如何添加文本其他特征信息到网络里呢，进而增加准确度？比如文本的位置信息

when i use the code，got TypeError: __call__() takes 2 positional arguments but 3 were given whta is the problem？ how to fix it ？ thinks

pip install tsai Successfully installed attrs-22.1.0 blis-0.7.8 catalogue-2.0.8 certifi-2022.6.15 charset-normalizer-2.1.1 click-8.1.3 cymem-2.0.6 fastai-2.7.9 fastcore-1.5.22 fastdownload-0.0.7 fastjsonschema-2.16.1 fastprogress-1.0.3 idna-3.3 imbalanced-learn-0.9.0 importlib-resources-5.9.0 jinja2-3.1.2 jsonschema-4.14.0 langcodes-3.3.0 llvmlite-0.39.0 murmurhash-1.0.8 nbformat-5.4.0 numba-0.56.0 pathy-0.6.2 pkgutil-resolve-name-1.3.10 preshed-3.0.7...

我自己跑了一下，为什么跑出来的效果这么差呢？ 训练到最后的日志日下： [->] step 696/792 loss 8.37 acc 0.37 [->] step 697/792 loss 17.91 acc 0.37 [->] step 698/792 loss 25.09 acc 0.37 [->] step 699/792 loss 13.42 acc 0.37...

模型跑长文本数据时loss波动太大，训练不出来，这是为什么呢？

6

感谢大佬分享！ 本人跑代码中的短文本时，结果和你发的一样，然后我就用模型跑长文本，还是THUCNews数据集，用的数据是：https://github.com/gaussic/text-classification-cnn-rnn中用到的数据，下载链接: https://pan.baidu.com/s/1hugrfRu 密码: qfud pad_size设置成600或1000都不行，日志如下： Loading data... 50000it [00:27, 1849.42it/s] 5000it [00:02, 1879.22it/s] 10000it [00:05, 1667.15it/s] Time usage: 0:00:36 Epoch [1/20] Iter: 0, Train Loss: 2.2, Train Acc: 17.97%,...

[BUG]: Repository Not Found for url: https://huggingface.co/pretrain/resolve/main/tokenizer_config.json

1

### 🐛 Describe the bug when i run ChatGPT: $ python train_reward_model.py --pretrain pretrain get : /home//anaconda3/envs/s20230216e310gpt/lib/python3.10/site-packages/torch/library.py:130: UserWarning: Overriding a previously registered kernel for the same operator and the same...

experiments/rationale_allenai-unifiedqa-t5-base_detr_QCM-LE_lr5e-05_bs16_op512_ep20/predictions_ans_eval.json

3

when i run : run_training.sh get ： $ bash run_training.sh args Namespace(data_root='data', output_dir='experiments', model='allenai/unifiedqa-t5-base', options=['A', 'B', 'C', 'D', 'E'], epoch=20, lr=5e-05, bs=8, input_len=512, output_len=512, eval_bs=4, eval_acc=10, train_split='train', val_split='val', test_split='test', use_generate=False,...

trainer.py--2823行，filtered_state_dict变量没有定义就直接用，是不是写错了？应该是state_dict这个变量吧？

1

### Is there an existing issue for this? - [X] I have searched the existing issues ### Current Behavior  你好，发现代码trainer.py--2823行，filtered_state_dict变量没有定义就直接用，是不是写错了？应该是state_dict这个变量吧？ ### Expected Behavior 还是什么原因呢？ ### Steps To Reproduce Python3.10...

RuntimeError: CUDA error: CUBLAS_STATUS_INVALID_VALUE when calling `cublasGemmStridedBatchedExFix( handle, opa, opb, m, n, k, (void*)(&falpha), a, CUDA_R_16F, lda, stridea, b, CUDA_R_16F, ldb, strideb, (void*)(&fbeta), c, CUDA_R_16F, ldc, stridec, num_batches, CUDA_R_32F, CUBLAS_GEMM_DEFAULT_TENSOR_OP)`

5

when i run : # Taiyi-Stable-Diffusion-1B-Chinese-v0.1 get : >>> import torch >>> from diffusers import StableDiffusionPipeline >>> torch.backends.cudnn.benchmark = True >>> pipe = StableDiffusionPipeline.from_pretrained("IDEA-CCNL/Taiyi-Stable-Diffusion-1B-Chinese-v0.1", torch_dtype=torch.float16) >>> pipe.to('cuda') >>> prompt =...

运行examples/sft/finetune.sh报错，请问用的环境版本分别是多少的呢？

6

首先，感谢分享！ 用的机器是单机多卡，8*A800，centos7，cuda12.2.2，python3.12.2, Operating System: CentOS Linux 7 (Core) CPE OS Name: cpe:/o:centos:centos:7 Kernel: Linux 3.10.0-1160.el7.x86_64 Architecture: x86-64 具体操作步骤如下： 1，pip3 install torch torchvision torchaudio，自动安装的torch-2.2.1 2，pip3 install deepspeed，自动安装的deepspeed-0.14.0 3，pip3 install transformers，自动安装的transformers-4.38.2 4，pip3...